The aim of this proposal is to improve the accuracy and consistency of measurements of absorbed dose for neutron radiation therapy by providing national dosimetry standards and improved data on neutron interactions with tissue and tissue-equivalent materials. Neutron radiation therapy is being clinically tested at several centers in the United States, and at about 25 centers worldwide. To maximize the chances of success of this radiation therapy modality, good physical dosimetry is meeded; and to facilitate exchange of therapy experience between institutions, the U.S. dosimetry standards base must be accurate and consistent with the international standards system. The technical approach will be to develop a tissue-equivalent (TE) ionization chamber coupled with a neutron-insensitive gamma-ray dosimeter as the working national standard. This method of neutron dosimetry will be compared against a TE calorimeter, and against neutron absorbed doses determined from monoenergetic neutron fluence measurement combined with calculated kerma factors to obtain a check between three independent methods of neutron absorbed dose determination. The monoenergetic neutron fluences are obtained from the NBS 3-MeV positive-ion Van de Graaff accelerator. Agreement between the 3 independent methods, and intercomparisons with the Bureau International des Poids et Mesures near Paris and with other national standards laboratories will assure the accuracy and consistency of the U.S. dosimetry standards for neutron therapy. The experimental standards program will be supported by a theoretical neutron dosimetry program aimed at improving the knowledge of physical data and correction factors needed to interpret the experimental measurements. Such data and factors include: secondary particle spectra, kerma factors, W (mean energy per ion pair), and theoretical neutron cross section data.